One of the biggest obstacles for efficient drug delivery is specific cellular targeting. Liposomes have long been used for drug delivery, but do not possess targeting capabilities. This limitation may be circumvented by surface coating of colloidal delivery systems with peptides, proteins, carbohydrates, vitamins, or antibodies that target cell surface receptors or other biomolecules. Each of these coatings has significant drawbacks. One idealized system for drug delivery combines stabilized "mini-protein" ligands with a colloidal delivery vehicle. Our prior studies have shown that peptide-amphiphiles, whereby both a peptide "head group" and a lipid-like "tail" are present in the same molecule, can be used to engineer collagen-like triple-helical or alpha-helical mini-proteins. The tails serve to stabilize the head group structural elements. These peptide-amphiphiles can be designed to bind to specific cell surface receptors with high affinity, while being minimally degraded. In preliminary studies, we have prepared liposomes using a melanoma targeting peptide-amphiphile ligand, and shown that these liposomes were stable, the peptide-amphiphile retained triple-helical structure, and an encapsulated fluorescent dye was selectively delivered to cells. In the present proposal we will develop peptide-amphiphile liposomes and polyPro dendrimers as a new class of drug delivery vehicles, taking advantage of melanoma receptor targeting and liposome activation by melanoma proteases. Transfer of drug into the cell will also be improved by incorporation of cell membrane traversing peptide-amphiphiles. A variety of biophysical and biological techniques will be used to evaluate peptide-amphiphile structure-function relationships. The selectivity of peptideamphiphile colloidal delivery vehicles will be analyzed using metastatic melanoma cells and other cell types (fibroblasts, endothelial cells) found in the melanoma microenvironment.